Eukaryotic genomes are filled with sequences derived by the reverse flow of information from RNA to DNA. Insertion of these reverse transcripts can induce spontaneous genetic disorders and cancer and have over evolutionary time completely reshaped eukaryotic genomes by generating enormous amounts of repetitive DNA. The most active, yet poorly understood, machinery responsible for these insertions is encoded by the non-long terminal repeat (non-LTR) retrotransposable elements One of the best characterized non-LTR elements is the R2 element of arthropods. This element exclusively inserts into the 28S rRNA genes of its host which greatly simplifies its characterization. The single protein encoded by R2 has remarkable specificity for both its DNA target site and RNA template enabling in vitro studies of its retrotransposition mechanism. This mechanism involves simple cleavage of the DNA target and the polymerization of the reverse transcript directly onto the cleaved chromosome, and thus is fundamentally different from the well-described retroviral mechanism of retrotransposition. Instead non-LTR elements are related to and their retrotransposition shares features with telomerase, the enzyme responsible for telomeres. The specific aim of this proposal is to fully characterize the R2 retrotranspostion mechanism as a model for all non-LTR elements. Detailed mutagenesis of the protein and nucleic acid components and characterization of the enzymatic activities of the protein are proposed. These in vitro studies will be complemented with in vivo studies of the R2 integration reaction in Drosophila melanogaster. Purified R2 protein/RNA when infected into D. melanogaster embryos can integrate foreign sequences into the 28S genes. This integration system will enable studies of the transcription and translation of R2 elements within the rDNA units of cell's nucleolus. These combined studies will address general questions about the mechanism of non-LTR retrotransposition, as well as questions concerning transcription regulation of the rRNA genes. Finally, a growing number of non-LTR elements have been characteriz4ed that retain different insertion specificities for unique sites in their host's genome. The enzymatic machinery encoded by these elements appears identical to that of R2. Our long term goal is to isolate and alter the target specific binding component of the integration reaction , which enable us to design systems that we will insert foreign sequences at other unique locations in eukaryotic genomes.